ORIGINAL RESEARCH article

Front. Microbiol., 09 January 2018

Sec. Evolutionary and Genomic Microbiology

Volume 8 - 2017 | https://doi.org/10.3389/fmicb.2017.02637

Genomic Analysis of Two Phylogenetically Distinct Nitrospira Species Reveals Their Genomic Plasticity and Functional Diversity

  • 1. Department of Life Science and Medical Bioscience, Waseda University, Tokyo, Japan

  • 2. Department of Material and Environmental Chemistry, Graduate School of Engineering, Utsunomiya University, Tochigi, Japan

  • 3. Bio-Measurement Research Group, Biomedical Research Institute, National Institute of Advanced Industrial Science and Technology, Ibaraki, Japan

Abstract

The genus Nitrospira represents a dominant group of nitrite-oxidizing bacteria in natural and engineered ecosystems. This genus is phylogenetically divided into six lineages, for which vast phylogenetic and functional diversity has been revealed by recent molecular ecophysiological analyses. However, the genetic basis underlying these phenotypic differences remains largely unknown because of the lack of genome sequences representing their diversity. To gain a more comprehensive understanding of Nitrospira, we performed genomic comparisons between two Nitrospira strains (ND1 and NJ1 belonging to lineages I and II, respectively) previously isolated from activated sludge. In addition, the genomes of these strains were systematically compared with previously reported six Nitrospira genomes to reveal their similarity and presence/absence of several functional genes/operons. Comparisons of Nitrospira genomes indicated that their genomic diversity reflects phenotypic differences and versatile nitrogen metabolisms. Although most genes involved in key metabolic pathways were conserved between strains ND1 and NJ1, assimilatory nitrite reduction pathways of the two Nitrospira strains were different. In addition, the genomes of both strains contain a phylogenetically different urease locus and we confirmed their ureolytic activity. During gene annotation of strain NJ1, we found a gene cluster encoding a quorum-sensing system. From the enriched supernatant of strain NJ1, we successfully identified seven types of acyl-homoserine lactones with a range of C10–C14. In addition, the genome of strain NJ1 lacks genes relevant to flagella and the clustered regularly interspaced short palindromic repeat (CRISPR)-Cas (CRISPR-associated genes) systems, whereas most nitrifying bacteria including strain ND1 possess these genomic elements. These findings enhance our understanding of genomic plasticity and functional diversity among members of the genus Nitrospira.

Introduction

Nitrification is a key aerobic process of the nitrogen cycle, which is catalyzed by chemolithoautotrophic ammonia-oxidizing archaea, ammonia-oxidizing bacteria, and nitrite-oxidizing bacteria (NOB). NOB converting nitrite to nitrate prevent toxic nitrite accumulation () and produce nitrate, which is an important source of nitrogen assimilation by many microorganisms and plants. Moreover, NOB contribute to the biological nitrogen removal process of wastewater treatment plants (WWTPs), preventing costal ecosystem eutrophication ().

To date, it has been reported that the chemolithoautotrophic NOB comprise seven genera (Nitrobacter, Nitrococcus, Nitrospina, Nitrospira, Nitrotoga, Nitrolancea, and Candidatus Nitromaritima) (). Of these NOB, the genus Nitrospira is globally distributed in soils, oceans, freshwater habitats, hot springs, and WWTPs (; ; ; ; ; ; ; ). The comparative analysis of 16S rRNA gene sequences revealed that the genus Nitrospira was classified into six different phylogenetic lineages (; ), which reflected the diversity of Nitrospira habitats (). Remarkably, Nitrospira lineages I and II were ubiquitously detected as dominant NOB in the activated sludge of WWTPs with competition and the partitioning of ecological niches between phylogenetically different populations (; ; ; ). However, Nitrospira is notoriously recalcitrant to isolation and representative Nitrospira strains from activated sludge were limited. Recently, the development of a separation technique using optical tweezers or a cell sorting system enabled the isolation of Nitrospira defluvii, Nitrospira lenta, Nitrospira sp. ND1, and Nitrospira japonica from WWTPs (; ; ), and their physiological properties and kinetic parameters were partially investigated (; ).

Recently, the first metagenomic analysis of N. defluvii belonging to lineage I revealed a novel nitrite-oxidoreductase, which is a key enzyme involved in nitrite oxidation by Nitrospira, and predicted that Nitrospira evolved from microaerophilic or anaerobic ancestors (). Subsequently, genomic analysis and physiological characterization of Nitrospira moscoviensis belonging to lineage II revealed growth of this organism by aerobic hydrogen oxidation () or the aerobic use of formate () without nitrite oxidation. Moreover, the wide distribution of environmental ureases and cyanases among Nitrospira was discovered by genomic analysis, suggesting that NOB could supply ammonia oxidizers lacking ureases and cyanases with ammonia and carbonate (; ). Surprisingly, it was reported that some Nitrospira species belonging to lineage II were able to perform complete ammonia oxidation (COMAMMOX), which oxidized ammonia to nitrate by a single microorganism (; ). In addition, four COMAMMOX and one canonical Nitrospira were recovered from metagenomes, and the comparative genomic analysis of COMAMMOX and related nitrifiers revealed their niche differentiation and the evolutionary history (Palomo et al., unpublished). These genomic analyses of Nitrospira have revealed their metabolic versatility, indicating that the ecophysiological role of Nitrospira is not exclusively nitrite oxidation.

As mentioned above, genomic analyses are powerful molecular approaches for finding unexpected metabolic versatility and predicting the evolution of unknown microorganisms. Although the complete genome sequences of representative Nitrospira strains belonging to lineages I and II, and several COMAMMOX Nitrospira genomes had previously been reported (; ; ; ; ; ; ), available information of Nitrospira genome sequences is still insufficient to understand their metabolism and evolution. In this study, we conducted genomic analysis of two Nitrospira strains, Nitrospira sp. strain ND1 and N. japonica strain NJ1, belonging to lineages I and II, respectively, both of which were previously isolated from activated sludge from a WWTP by our research group (; ), and compared their genomic information with those of previously reported Nitrospira strains. Our findings revealed remarkable heterogeneity in the genes involved in quorum-sensing systems, bacterial flagella and CRISPR-Cas system, and indicated that the phylogenetic diversity of Nitrospira correlated to adaptation to various environments.

Materials and Methods

Genome Reconstruction and Annotation

Nitrospira sp. strain ND1 and N. japonica strain NJ1 were isolated from activated sludge from a WWTP as previously reported (; ). DNA was extracted from strain ND1 or strain NJ1 using a NucleoSpin® Tissue DNA extraction kit (Takara Bio, Otsu, Japan) according to the manufacturer’s instructions. DNA sequencing for generating the genomes of the strains was performed by National Institute of Advanced Science and Technology. Briefly, paired-end (300 to 1,000 bp inserts, Nextera XT indexed) and Nextera mate-pair (1 to 14 kbp insert) libraries were generated and sequenced on an Illumina MiSeq instrument using V2 chemistry (2-bp × 250-bp reads). Raw reads were merged with SeqPrep using default settings, including the removal of sequencing adapters. Reads that failed to merge were quality trimmed and filtered using Nesoni v0.112. Both merged and processed single reads were assembled using SPAdes version 2.5.0 (), followed by manual curation of the assembly (). The reconstructed genome sequences of strains ND1 and NJ1 were deposited in the European Nucleotide Archive (ENA) under the accession numbers FWEX01000001-FWEX01000006 and LT828648 respectively.

The reconstructed draft and complete genomes of the strains ND1 and NJ1, respectively, were integrated into the MicroScope platform (), and associated annotations are publicly available in MicroScope (#17KHXP and #3D7KKV respectively). Coding sequences (CDS) were automatically predicted and annotated by using the MicroScope platform. In addition, function of the annotated CDS as hypothetical protein by the MicroScope platform was speculated by searching their similar proteins with BLAST in NCBI Reference proteins database. The predicted CDS were summarized into a table according to their predicted function, which included nitrogen metabolism, carbon fixation, respiration, and motility (Supplementary Data Sheet S2). CRISPR repeat sequences were identified using CRISPRfinder ().

Comparison of Nitrospira Genome Sequences

The average nucleotide identity (ANI) and the average amino acid identity (AAI) among Nitrospira genome sequences were calculated using online tools (). The parameters of the ANI calculation were adjusted to the following default settings: a minimum alignment length, a minimum identity, a fragment window size, and a fragment step size of 700 bp, 70%, 1,000 bp, and 200 bp, respectively. The minimum identity of the AAI calculation was adjusted to 30%. Homologous proteins between two Nitrospira strains were identified by using the phyloprofile exploration tool of MicroScope (). The parameters of the phyloprofile exploration were adjusted to the following default settings: a minimum alignment length and a minimum identity of 80 and 30%, respectively. In addition, based on gene annotation information of the Nitrospira genomes, the presence/absence of several functional genes/operons among the genomes was summarized into a figure (Figure 1).

FIGURE 1

Phylogenetic Analysis

Reference 16S rRNA and protein sequences for phylogenetic inference were obtained from the NCBI database. Multiple sequence alignments of these sequences were generated automatically using ClustalW2 (). Phylogenetic trees based on 16S rRNA gene sequences were constructed using MEGA 6 software with the Maximum Likelihood algorithm (). Also, phylogenetic trees were constructed for the amino acid sequences of selected proteins using the Maximum Likelihood algorithm.

Identification of AHLs in Nitrospira japonica Strain NJ1

Preparation for identification of AHLs using the LC–MS/MS was conducted according to previous studies (; ). First, the 1 L batch cultures of strain NJ1 were acidified to pH < 2 with HCl to avoid lactonolysis. Then the acidified samples were extracted five times with 200 ml of ethyl acetate. The extracted samples were dried down using an evaporator and recovered with 1 mL dimethylsulfoxide. Then, according to the previous study (; ), identification of the extracted AHLs was performed using a triple quadrupole/linear ion trap instrument (QTRAP5500, AB Sciex, Framingham, MA, United States) with an electrospray ionization source coupled to an UHPLC system (Nexera X2, Shimadzu, Kyoto, Japan). Chromatographic separation was achieved on a C18 column (Kinetex F5, Φ 2.1 mm × 150 mm, 2.6 μm, Phenomenex, Torrance, CA, United States). MS/MS spectra were recorded as reported previously (). Precursor ion scanning experiments were performed in positive ion mode for analysis. In the experiments, one quadrupole (Q1) was set to scan a mass range of m/z 80 to 500 Da and other quadrupole (Q3) was used for detection of the product ion at m/z 102, which indicates presence of lactone ring in the compound (). AHLs were identified by precursor ion scanning experiments and comparison with synthetic AHL standards. AHL standards, C10-HSL, 3-oxo-C10-HSL, C12-HSL, 3-oxo-C12-HSL, 3-OH-C12-HSL, C14-HSL, and 3-oxo-C14-HSL, were synthesized using a previously described method (). Purity of these AHL standards was checked by using nuclear magnetic resonance (NMR).

Results and Discussion

Genomic Comparison of Nitrospira Cultures

The final high-quality draft genome assembly of the ND1 genome consisted of six contigs. The largest and N50 contig size was 3,463,908 bp, and the total assembly size was 4,454,928 bp for the ND1 genome. The obtained genome sequence of strain ND1 contained 4,612 CDS (Supplementary Table S1). By contrast, the obtained sequence reads from strain NJ1 were assembled into one scaffold. The reconstructed complete genome sequence of strain NJ1 was 4,084,817 bp in length and contained 4,150 predicted CDS (Supplementary Table S1). These two Nitrospira strains appeared to contain no plasmids.

We compared the two reconstructed genome sequences with the published complete genome sequences of N. defluvii (NC_014355), N. moscoviensis (NZ_CP011801), and Nitrospira inopinata (NZ_LN885086), and published contig sequences of Candidatus Nitrospira nitrosa (CZQA01000001_CZQA01000015) and Candidatus Nitrospira nitrificans (CZPZ01000001_CZPZ01000036), based on 16S rRNA gene similarity, ANI, AAI, and the presence/absence of several functional genes. Phylogenetic analysis based on the 16S rRNA gene revealed that strain ND1 and the N. defluvii strain belonged to lineage I, and the other five Nitrospira strains belonged to lineage II (Figure 1). Strain ND1 shares a similar 16S rRNA gene sequence to N. defluvii (DQ059545; 99.8%) (). Strain NJ1 is distantly related to N. moscoviensis (CP011801; 96.1%) () and N. lenta (KF724505; 96.1%) (), and is also distantly related to strain ND1 (92.6%). The ANI and AAI among Nitrospira genome sequences indicated a close genetic relatedness between strain ND1 and N. defluvii (91.4%, Figure 1A). But, the genome sequence of strain ND1 contains genes encoding cytochrome c nitrite reductase (NrfA) and a urease locus, neither of which is contained in the genome of N. defluvii (Figure 1B). By contrast, the ANI and AAI among the genome sequence of Nitrospira strains belonging to lineage II were below 80%, suggesting a putative functional diversity due to the evolutionary distance of the different Nitrospira lineage II genomes (Figure 1A). Also, the strain ND1 genome contained 2,717 out of 4,612 CDS as homologous genes shared by the strain NJ1 genome, suggesting that their genomes contained a considerably different set of genes despite being isolated from the same activated sludge (Figure 1A).

The CDS annotated with unknown function of the genome sequence in the strains ND1 and NJ1 accounted for 40.5 and 41.5%, respectively. Based on the annotated CDS with certain function in the genome sequence of the two strains, their key metabolic pathways were predicted and compared with other Nitrospira strains. Comparing genomic features among Nitrospira strains, we first focused on the genes involved in nitrogen metabolism, such as nitrite reduction and urea degradation. Next, from the strain NJ1 genome, we discovered the key genes of acyl-homoserine lactone (AHL)-type quorum-sensing (QS) systems, which was reported in the previous study (; ), and we identified the chemical structures of the AHLs as autoinducers of strain NJ1. Moreover, the genome sequence of strain NJ1 lacks genes encoding bacterial flagella and CRISPR-Cas system, which were reported in genomic analyses of other Nitrospira (Figure 2).

FIGURE 2

Nitrogen Metabolism of Strains ND1 and NJ1

Nitrospira is a chemolithoautotrophic nitrite-oxidizing bacterium, and it was confirmed that strains ND1 and NJ1 were able to grow in mineral medium with oxidizing nitrite to nitrate in the previous reports (; ). Both strains appeared to oxidize nitrite for energy conservation and to assimilate nitrite as a nitrogen source. Although most genes involved in key metabolic pathways (nitrite oxidation, the electron transport chain, glycolysis/gluconeogenesis, the tricarboxylic acid cycle, the pentose phosphate pathway, and sulfur assimilation) were conserved between the two Nitrospira genome sequences, genes involved in nitrogen assimilation differed (Figure 2 and Supplementary Data Sheet S2). The genome sequence of strain NJ1 contained genes encoding an Octaheme cytochrome c nitrite reductase (ONR) and a cytochrome bc complex (Figure 2). Previously, it was reported that nitrite reduction in N. moscoviensis was most likely catalyzed by ONR (). Meanwhile, the genome sequence of strain ND1 contained two nitrite reductase genes, nirA and nrfAH (Figure 2). NirA is reported to be a cytoplasmic ferredoxin-dependent nitrite reductase for nitrogen assimilation in plant cells and prokaryote cells (; ; ). By contrast, NrfAH is a periplasmic cytochrome c nitrite reductase using electrons from quinol, not for assimilation, but rather for respiratory ammonification (). Previous studies reported that nitrite reduction in N. defluvii and N. lenta was most likely catalyzed by NirA (; ), whereas reported the presence of NrfAH in the genome of N. inopinata, but nitrite reduction by NrfAH was not shown.

Comparing the genomic region of genes involved in nitrogen metabolism among Nitrospira strains, the gene encoding ONR in strain NJ1 and N. moscoviensis, and the nirA gene in strain ND1 and N. defluvii are located close to the glnA gene encoding glutamine synthetase, the amtB genes encoding an ammonium transporter, and the glnK gene encoding P-II protein, which are known to be involved in nitrogen assimilation (Figure 3). Thus, nitrite reduction in strains ND1 and NJ1 was most likely catalyzed by NirA and ONR, respectively. Indeed, expression of the gene encoding ONR of strain NJ1 was confirmed by reverse-transcription (RT)-PCR (Supplementary Figure S1 and Supplementary Table S2). By contrast, the nrfAH gene (NSND_62628, NSND_62627, NITINOP_0678, and NITINOP_0679) in the genome sequence of strain ND1 and N. inopinata was not located close to the glnA gene (NSND_60567 and NITINOP_1778). Likely, their NrfAH was not used for assimilatory nitrite reduction. The genome sequence of strain ND1 and N. inopinata contain fdhABC gene (NSND_50216-NSND_50218) encoding formate dehydrogenase and hydABC gene (NITINOP_0586-NITINOP_0588) encoding hydrogenase respectively, which are able to use formate and hydrogen as electron donor of respiratory ammonification (). Besides, it was reported that Nitrospira strains grow with formate or hydrogen as substrate (; , ). Although we did not check their respiratory ammonification with formate and hydrogen, the NrfAH shared among Nitrospira genomes might be required for respiratory ammonification.

FIGURE 3

Recently, it was reported that some Nitrospira strains performed not only nitrite oxidation, but also urea degradation, cyanate degradation, and ammonia oxidation (; ; ; ). Our reconstructed genome sequences of strains ND1 and NJ1 also contained genes involved in the degradation of urea and cyanate (Figure 2), and we confirmed ureolytic activity of both strains (Supplementary Figure S2). Phylogenetic analysis based on the urease alpha subunits (UreC) revealed that the UreC proteins of most Nitrospira species, except strain NJ1, were affiliated with one Nitrospira clade (Figure 4). In addition, although previously reported Nitrospira genomes and strain ND1 lack the ureE gene encoding a urease accessory protein, the strain NJ1 genome sequence contained a complete gene cluster encoding urease accessory proteins including the UreE protein (Figure 3). Besides, the genomes of both strains contained the high affinity urea ABC transporter (urtABCDE) shared among most Nitrospira genomes possessing a urease locus (; ; ; ; Palomo et al., unpublished), and the strain NJ1 genome contained the Yut-like urea transporter, which is a single component channel for facilitated diffusion of urea () and was not contained in other reported Nitrospira genomes.

FIGURE 4

Therefore, the phylogeny of UreC proteins and the presence of the ureE gene and a different urea transporter indicated that the ureolytic machinery of strain NJ1 differs from that of other Nitrospira species. Besides, in the phylogenetic tree based on UreC, the Nitrospira clade included not only the UreC of N. moscoviensis, N. lenta, N. inopinata, Ca. Nitrospira nitrosa, and Ca. Nitrospira nitrificans belonging to Nitrospira lineage II, but also the UreC of strain ND1 belonging to Nitrospira lineage I. Although the wide distribution of environmental ureases among Nitrospira was reported in the previous study (), the reported and published Nitrospira genomes possessing a urease belong to only lineage II. So, the possession of the ureC gene in the genome of strain ND1 belonging to Nitrospira lineage I indicated that urease genes widely distribute among Nitrospira, not only lineage II.

The Quorum-Sensing System of N. japonica Strain NJ1

Unexpectedly, the genome sequence of strain NJ1 contained the njpI gene (NSJP_1610) and the njpR gene (NSJP_1611) encoding a putative synthase and a receptor of autoinducers, respectively, which are known signaling compounds of QS systems (Figure 2 and Supplementary Data Sheet S2). The QS system is a cell–cell communication mechanism among bacteria in environments that employs autoinducers to regulate bacterial behaviors such as biofilm formation, bacterial motility, luminescence, and plasmid transfer (; ). AHLs, common signaling compounds of QS systems, are synthesized by the LuxI protein as autoinducer synthases. Previously, genes encoding a LuxI homolog and AHLs were identified in the genome sequences of Nitrosomonas europaea and Nitrosospira multiformis, which are known ammonia-oxidizing bacteria, and Nitrobacter winogradskyi, which is known as NOB (; ; ). In addition, a metagenomic clone from the phylum Nitrospirae was reported to possess AubI/AubR as a LuxI/LuxR homolog, producing N-dodecanoyl-L-homoserinelactone (C12-HSL) as an autoinducer (). Recently, it was reported that the genome sequence of N. moscoviensis, N. inopinata, and Ca. Nitrospira nitrificans contains a gene encoding AHL synthase homologs (). Phylogenetic analysis based on the amino acid sequence of autoinducer synthases revealed that the LuxI homologs of Nitrospira were each clustered into single clades (Figure 5), and that NjpI, a LuxI homolog in strain NJ1, was most closely related to that of Ca. Nitrospira nitrificans (67.5%, CUS35775). But, since no gene encoding a LuxI homolog was detected in the genome sequence of strain ND1 or N. defluvii belonging to lineage I, the LuxI homologs might be partially distributed among the genomes of Nitrospira species.

FIGURE 5

Identification of AHLs in Strain NJ1

We investigated AHL production in the culture supernatant of strain NJ1 using an AHL bioassay with Chromobacterium violaceum strain VIR07 (). We extracted AHLs from the batch culture of strain NJ1 according to the reported protocol (). Using the AHL bioassay, AHLs were successfully detected from the enriched supernatant (Supplementary Figure S3). Subsequently, we identified the AHLs in the enriched supernatant using an LC–MS/MS. Unexpectedly, LC–MS/MS analysis revealed that seven types of AHLs, namely N-decanoyl-L-homoserine lactone (C10-HSL), N-(3-oxodecanoyl)-L-homoserine lactone (3-oxo-C10-HSL), N-dodecanoyl-L-homoserine lactone (C12-HSL), N-(3-hydroxydodecanoyl)-L-homoserine lactone (3-OH-C12-HSL), N-(3-oxododecanoyl)-L-homoserine lactone (3-oxo-C12-HSL), N-tetradecanoyl-L-homoserine lactone (C14-HSL), and N-(3-oxotetradecanoyl)-L-homoserine lactone (3-oxo-C14-HSL) were present in the supernatant (Figure 6 and Supplementary Data Sheet S3). Comparison among relative abundance of these AHLs based on peak area of each compounds suggested that C12-HSL was the dominant autoinducer (Supplementary Table S3). Interestingly, although the genome sequence of strain NJ1 contains one njpI gene encoding an AHL synthase, the supernatant contained seven types of AHL. Previously, reported that several kinds of AHLs were identified from recombinant Escherichia coli cells containing the nwiI gene encoding an AHL synthase of N. winogradskyi. In addition, it was often reported that several types of AHLs were biosynthesized by only one AHL synthase from some bacteria, e.g., Pseudomonas putida WCS358 (). It was also accepted that AHL synthase preferred to use not only one specific acyl carrier protein (ACP) charged with fatty acids, but also closely related acyl-ACPs, when AHLs were synthesized from S-adenosylmethionine and acyl-ACP (). Thus, the AHL synthase of strain NJ1 likely preferred to synthesize AHLs within a range of C10–C14. Comparing the chemical structures of the identified AHLs of strain NJ1 with those of nitrifying bacteria in previous studies (Supplementary Table S4), strain NJ1 appeared to synthesize and secrete different types of AHLs from other nitrifying bacteria. In addition, its dominant autoinducer, 12-HSL, was similar to that of a metagenomic clone from the phylum Nitrospirae (). Thus, it was likely that partial Nitrospira strains might produce autoinducers conserved among Nitrospira species.

FIGURE 6

In recent studies, it was suggested that nitrogen metabolism and the flux of nitrogen oxide in N. winogradskyi were regulated by AHL signaling (; ). Although the role of AHL-based QS system in strain NJ1 is still unknown, future transcriptional analyses using nitrifying pure cultures possessing QS systems may help to explain the mechanisms of transcriptional regulation and interspecies communication mediated by AHLs.

Bacterial Flagella and Chemotaxis of Strain ND1

The genome sequence of strain ND1 contained almost all of the known flagella genes such as flg, flh, fli, and motAB, and the principal chemotaxis genes such as che and MCP (Supplementary Data Sheet S2). However, the strain ND1 genome curiously lacked the flhDC gene as a master transcriptional regulator of flagellation and chemotaxis (). Likewise, all Nitrospira genome published on the NCBI database also lacks the flhDC gene. Following genome analysis of N. winogradskyi, a canonical NOB, it was reported that the flhDC gene was lacking, and flagellation and chemotaxis in N. winogradskyi appeared to be differently regulated by other genes (). Interestingly, the genome of strain ND1 and N. defluvii contained a homologous gene encoding the transcription factor FleQ and the two-component system FleSR similar to Pseudomonas aeruginosa (; ). It was also reported that FleQ activates the transcription of genes involved in flagellar biogenesis and the FleSR two-component system depending on σ54 (). Thus, flagellation and chemotaxis of strain ND1 and N. defluvii appeared to be controlled by FleQ as a master transcriptional regulator.

Lack of Flagella in N. japonica Strain NJ1

Only two genes involved in bacterial flagella were encoded in the N. japonica strain NJ1 genome sequence, suggesting that strain NJ1 lacks the potential for motility using flagella (Figure 2). As mentioned above, the genome sequence of strain ND1 contained almost all of the genes involved in bacterial flagella. Likewise, genes involved in flagella were reported in genomic analyses of other Nitrospira (; ), and flagella of N. inopinata were recently observed by transmission electron microscopy (). Moreover, genes encoding bacterial flagella have been reported in the genome sequences of the genera Nitrobacter, Nitrococcus, Nitrospina, Nitrosomonas, and Nitrosospira, which are known nitrifying bacteria (; ; ; ; ). Strain NJ1 is therefore unusual among nitrifying bacteria in that it lacks genes encoding bacterial flagella. Interestingly, it was reported that Nitrospira cells in strain NJ1 formed micro-colonies in the same manner as strain ND1 (; ). Thus, it is likely that the potential for motility using flagella is not associated with micro-colony formation.

CRISPR-Cas System of Nitrospira Genomes

The genome sequence of strain ND1 contained genes encoding the clustered regularly interspaced short palindromic repeat (CRISPR)-Cas (CRISPR-associated genes) systems, similar to those detected in N. defluvii and N. moscoviensis (; ) (Figure 2). By using CRISPRfinder (), it was confirmed that the genome sequence of strain ND1 contained CRISPR including 31 spacer sequences (crRNA) between the repeat units respectively (Supplementary Data Sheet S2). By contrast, the genome sequence of strain NJ1 contained no genes encoding the CRISPR-Cas system (Figure 2). The CRISPR-Cas system is a widely distributed prokaryotic immune system among bacteria and archaea. Likely, strain ND1 was equipped for resistance against phage using its own CRISPR-Cas system in the same manner as other Nitrospira isolates.

Genomic Comparison between Strain ND1 and Nitrospira defluvii

Although several Nitrospira genomes were reported in the previous studies, the genome sequence of Nitrospira strains belonging to lineage I was only reconstructed from N. defluvii. Since the reported strain ND1 genome in this study was the second genome sequence of Nitrospira belonging to lineage I, homologous and non-homologous proteins between the two Nitrospira strains were identified by using the phyloprofile exploration tool of MicroScope (). The genome sequence of strain ND1 contained 3,577 genes encoding homologous and 1,047 genes encoding non-homologous proteins with N. defluvii. The latter included the urease locus and the nrfAH gene mentioned above. In addition, genes encoding superoxide dismutase and catalase were also identified as non-homologous genes with N. defluvii. These two proteins are known as key enzymes for the defense against reactive oxygen species (ROS), and were also contained in the genome sequence of N. moscoviensis (). By contrast, the genome sequence of N. defluvii contained 752 genes encoding non-homologous proteins with strain ND1, which included the aoxAB gene encoding large and small subunits of arsenite oxidase (AOX). It was reported that AOX functions in arsenite detoxification or enables N. defluvii to use arsenite as electron donor ().

Recently, genomics of Nitrospira strains belonging to lineage II revealed that Nitrospira bacteria possess hugely diverse functions such as utilization of formate or hydrogen, ammonia production from urea and cyanate, and ammonia oxidation (, ; ; ). By contrast, due to insufficient genomic information, it has been unknown whether Nitrospira strains belonging to lineage I possess diverse functions. Owing to novel genomic information of strain ND1 revealed in this study, it is expected that Nitrospira lineage I possess genomic plasticity in the same manner as Nitrospira lineage II. Thus, further genomic information from Nitrospira strains belonging to lineage I and physiological characterization using available pure strains may help to illuminate still unknown functions of Nitrospira lineage I.

Conclusion

In this study, the genome sequences of strains ND1 and NJ1 were reconstructed, and compared with those of other Nitrospira strains. Comparison of the genomes among Nitrospira lineages I and II revealed diversity of genes involved in urea degradation. Also, the comparative genomics illuminated the presence/absence of several functional genes/operons, such as motility, CRISPR-Cas system and QS system. In addition, although strains ND1 and NJ1 have been isolated from the same activated sludge, significant functional differences were predicted from their genomic information. These functional differences among the Nitrospira strains may be important factors in ecological niche differentiation among these bacteria in the activated sludge. Remarkably, we detected a gene encoding an AHL synthase involved in a QS system in the genome sequence of strain NJ1, and identified seven types of AHL within a range of C10–C14 from the supernatant of strain NJ1. Future transcriptional analyses using nitrifying pure cultures possessing QS systems may help to explain the mechanisms of transcriptional regulation and interspecies communication mediated by AHLs.

Statements

Author contributions

NU, HF, and ST designed the experiments. YSe performed and wrote the genome sequencing and assembly. NU performed the genome analysis. NU performed the gene expression experiments. NU, HF, and ST analyzed the data. YSh and TM identify acyl-homoserine lactones from Nitrospira japonica strain NJ1. NU drafted the manuscript. NU, HF, YSe, and ST read and edited the manuscript.

Funding

This work was supported by Japan Science Society, Kurita Water and Environment Foundation, Japan Society for the Promotion of Science (Award ID: JP 15J08166), Institute for Fermentation, Osaka, and Waseda University.

Acknowledgments

The authors thank Masaru Jinno for help with the ureolytic activity test for the two Nitrospira isolates and for providing biomass of Nitrospira pure cultures for genomic analyses. They also thank the LABGeM team at Genoscope, France, for annotation support within the MicroScope platform.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2017.02637/full#supplementary-material

References

Summary

Keywords

Nitrospira, nitrification, genome, autoinducer, activated sludge

Citation

Ushiki N, Fujitani H, Shimada Y, Morohoshi T, Sekiguchi Y and Tsuneda S (2018) Genomic Analysis of Two Phylogenetically Distinct Nitrospira Species Reveals Their Genomic Plasticity and Functional Diversity. Front. Microbiol. 8:2637. doi: 10.3389/fmicb.2017.02637

Received

28 September 2017

Accepted

18 December 2017

Published

09 January 2018

Volume

8 - 2017

Edited by

Martin G. Klotz, Washington State University Tri-Cities, United States

Reviewed by

Sebastian Luecker, Radboud University Nijmegen, Netherlands; Hanna Koch, University of Oldenburg, Germany

Updates

Copyright

*Correspondence: Satoshi Tsuneda,

This article was submitted to Evolutionary and Genomic Microbiology, a section of the journal Frontiers in Microbiology

Disclaimer

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Outline

Figures

Cite article

Copy to clipboard


Export citation file


Share article

Article metrics